Bia 1981

Download as pdf or txt
Download as pdf or txt
You are on page 1of 15

Symposium on Body Fluid and Electrolyte Disorders

Mixed Acid Base Disturbances: A


Clinical Approach

Margaret Bia, M.D., *


and Samuel O. Thier, MD. t

A mixed acid-base disturbance is defined as the simultaneous coexistence


of two or more simple disorders in the same patient. In any discussion of
mixed acid-base disorders, it is crucial to understand and employ the
appropriate tenninology. The suffix "osis" (acidosis, alkalosis) refers to the
underlying metabolic or respiratory process while the suffix "emia" (acide-
mia, alkalemia) refers to the net effect of one or more of these processes on the
blood pH. A patient may therefore have several processes appropriately
termed acidosis and/or alkalosis, but the net result can only be an acidemia
OR an alkalemia OR neither. The latter would occur when the differing
effects of the underlying processes on the blood pH exactly offset each other.
The blood pH resulting from a mixed acid-base disturbance can be severely
deranged as, for example, when metabolic acidosis coexists with respiratory
acidosis, or may be normal as, for example, when metabolic acidosis is present
with metabolic alkalosis. Mixed acid-base disturbances are most commonly
predictable on the basis of the clinical setting and physical examination;
laboratory data serve mainly to confinn the clinical impression.
Proficiency at analyzing mixed' acid-base disorders is a skill worth
mastering not because the analysis is an interesting academic exercise, but
because it is often of considerable diagnostic and therapeutic value. For
instance, in a febrile patient with alkalemia due to a respiratory alkalosis from
sepsis, a widening of the anion gap and decrease in serum bicarbonate level
may indicate the development of a lactic acidosis and may provide the earliest
evidence of impending difficulties from septic shock. Similarly, in a patient
with a severe metabolic acidosis and a "normal" Pco 2 , recognition that the
"nonnal" Pco 2 in this setting is inappropriately high for metabolic acidosis
and therefore represents a second primary disturbance would be critical to the
patient's management. In this case, the diagnosis of a primary respiratory
acidosis allows the physician to focus on the severity of the pulmonary

'Assistant Professor of Medicine; Assistant Director, Dialysis and Transplant Program, Yale
University School of Medicine, New Haven, Connecticut
tDavid Paige Smith Professor of Medicine; Chairman, Department of Medicine, Yale University
School of Medicine, New Haven, Connecticut

Medical Clinics of North America - Vo!. 65, No. 2, March 1981 347
348 MARGARET BIA AND SAMUEL O. THIER

compromise and initiate appropriate treatment to lower the Pco 2 and compen-
sate for the metabolic acidosis. Identification of mixed disturbances has other
important therapeutic implications as well. Patients with chronic respiratory
acidosis due to severe pulmonary disease often have a coexisting metabolic
alkalosis secondary to the aggressive use of diuretics to treat congestive heart
failure. a7 Failure to appreciate and therefore treat the alkalosis, particularly
when the patients are on respirators, can add to the respiratory depression a
and can create difficulty in weaning such patients from ventilatory support.
Thus it should be clear that an understanding of complex acid-base disturb-
ances provides the clinician with an important diagnostic skill that can be
utilized to provide better evaluation and management.

INTRODUCTORY CONCEPTS

The pH of extracellular fluid is defined by the Henderson-Hasselbalch


equation:

HCOa
pH = pK log .03 X PC02 (1)

Alternatively, the hydrogen ion concentration of the extracellular fluid can be


expressed more directly as:

H + =24 PC02 (2)


X HCOa

The latter expression, derived by Kassirer et al.,22 is clinically more useful


since it allows the rapid calculation of the expected hydrogen ion concentra-
tion without the use of a logarithm table.
Evident in both equations is the concept that the final hydrogen ion
concentration or pH is determined not by the absolute amount of either HCOa
or Pco2 but by the ratio of the two. Thus a normal HCOa concentration does
not necessarily imply that the pH is normal, since the Pco 2 could be elevated
and result in a significant depression in blood pH. This situation could occur
when a respiratory acidosis coexists with a metabolic acidosis. Similarly, an
abnormal HCOa concentration does not necessarily indicate an abnormal pH
since the Pco 2 can be abnormal to proportionally the same extent. The result
would be a normal ratio of Pco 2 /HCO a and therefore a normal hydrogen ion
concentration and pH. This picture could be produced by a metabolic
acidosis, depressing the serum HCO a, occurring with a primary respiratory
alkalosis that depresses the Pco 2 even further. Finally, it must be pointed out
that even when all three measurements, pH, Pco 2 , HCO a, are normal,
abnormalities in acid-base balance may still be present. For example, if a
metabolic acidosis (from renal failure) coexists with a metabolic alkalosis of
equal magnitude (from vomiting), all three measurements can fall within the
normal range. Thus, a full comprehension of acid-base disturbances demands
a careful analysis both of the acid-base indices and the clinical processes
occurring in the patient.
MIXED ACID BASE DISTURBANCES 349
Although the normal pH ranges between 7.36 and 7.44, the value in a
given individual varies much less than this. A pH of 7.40 corresponds to a
hydrogen ion concentration of 40 nEq/L. Between the range of pH = 7.20 to
7.50, each change in pH of 0.10 units corresponds to a change in hydrogen ion
concentration of 10 nEq/L.22 Knowledge of this linear relationship along with
the use of Equation 2 allows the clinician to make a quick check of the
internal consistency of the blood gas values. Thus the reported values for
Pco 2 and HC03 from a blood gas determination can be utilized to calculate
the expected hydrogen ion concentration which can then be converted into
pH units according to the above relationship. Failure of this calculated pH to
agree with the measured pH implies that one or more of the reported values is
inaccurate since pH, Pco 2 and HC0 3 - must always relate to each other
according to the Henderson-Hasselbalch equation.

LIMITS OF COMPENSATION

Primary respiratory acid-base disturbances invoke secondary metabolic


responses and primary metabolic disorders induce secondary respiratory
responses. Since the definition of a simple disturbance includes both the
initial process producing a change in HC0 3 and Pco 2 and all of the compensa-
tory mechanisms affecting these substances,4o lack of appropriate compensa-
tion for a simple disturbance is evidence for a mixed disturbance. Therefore,
it is critical to know both the magnitude and the time course of the
compensatory responses to simple disorders to identify correctly the presence
of a mixed disturbance. A detailed analysis of the compensatory responses in
simple acid-base disorders is presented by Narins and Gardner. For the
present discussion, only the key features of these compensatory responses
will be outlined with emphasis placed on the use of this knowledge to
diagnose a mixed acid-base disturbance. Major points to remember in analyz-
ing the secondary changes in a given disturbance are: (1) compensation rarely

Table 1. Compensatory Responses in Simple Acid-Base Disturbances


CLINICAL GUIDE FOR COMPENSATION LIMIT

Metabolic acidosis pCO, = 1.5 X HC0 3 + 8.4 10 mmHg

Metabolic alkalosis Each mEq t in HC0 3 --> 0.5-1.0 mmHg t in PCO, 55 mmHg

Acute Respiratory HCO, t by 3-4 mEq/L 30 mEq/L


acidosis

Chronic respiratory Each mmHg t in PCO, --> 0.4 mEq/L t in HCO, 45 mEq/L
acidosis

Acute respiratory HC0 3 t by 2-4 mEq/L 18 mEq/L


alkalosis

Chronic respiratory Each mmHg t in PCO, -->0.5 mEq/L t in HC0 3 12-15 mEq/L
alkalosis
350 MARGARET BIA AND SAMUEL O. THIER

corrects the pH back to normal (except in chronic respiratory alkalosis where


compensation may be complete) and (2) compensation never over corrects.
Clinical guidelines for assessing the adequacy of the compensation and
the limits of compensation are listed in Table 1. In a metabolic acidosis, the
ventilatory response is rapid even though it may take about 12 hours to reach a
maximum in some cases. 34 The adequacy of hyperventilation (which can
lower Pco 2 down to a limit of approximately 10 mm Hg36) can be evaluated
using the formula derived by Albert, Dell, and Winters: 1

Pco 2 = 1.54 X HCOi + 8.36 ± 1.1 (3)

According to this equation, which should only be used in patients with a


metabolic acidosis, a Pco 2 higher than that predicted indicates inappropriate
CO 2 retention. A Pco 2 lower than the predicted value represents hyperven-
tilation out of proportion to that expected for respiratory compensation of a
metabolic acidosis. For example, in a diabetic with ketoacidosis producing a
decrease in serum HC03- to 12 mEq/L, a Pco 2 of 25 mm Hg (resultant pH =
7.30) would be an appropriate respiratory response. If, however, the same
patient had a PC0 2 of 32 (resultant pH = 7.18) a concomitant respiratory
acidosis should be diagnosed, whereas a PC0 2 of 20 (resultant pH 7.40)
would provide evidence for a coexistent respiratory alkalosis. The relation-
ship between HC0 3 and PC0 2 during metabolic acidosis as defined by this
equation is worth remembering since it can be easily applied at the bedside to
establish that a metabolic acidosis is simple or is complicated by a concomi-
tant primary respiratory disturbance. It must be remembered, however, that
since it can take 12 hours for the ventilatory response to reach a maximum or
to resolve once the underlying acidosis is corrected,34, caution must be
exercised in applying this formula in situations in which the acid-base
disturbance is changing rapidly.
The ventilatory response to a metabolic alkalosis is less efficient and
less predictable than the ventilatory response to a metabolic acidosis. 9 ,20
Although CO 2 is retained during a metabolic alkalosis, the magnit~de of the
response is not as great as is the fall in Pco 2 during metabolic acidosis.
Furthermore, the limit of compensatory CO 2 retention is reached at approxi-
mately 55 to 60 mm Hg,5, 10, 15,20 after which there is usually no further rise in
Pco 2 despite further increases in serum HC0 3-. In general, the Pco 2 rises by
0.5 to 1.0 mm Hg for every 1 mEq/L increment in serum HC0 3 concentra-
tion. 5, 9, 15, 31 Since the amount of CO 2 retained in response to metabolic
alkalosis is so variable, no single equation has been proven adequate to
describe the ventilatory response in all cases of metabolic alkalosis. It is
therefore best to assess the appropriateness of compensation according to the
limits of compensation. For instance, in a patient with a serum HC0 3- of 42
mEq/L due to metabolic alkalosis from nasogastric suction, a Pco 2 of 52 mm
Hg (resultant pH = 7.52) would be appropriate. In the same patient, a Pco 2 of
40 mm Hg (resultant pH 7.63) would clearly be an inappropriate response and
would therefore indicate a coexistent respiratory alkalosis. By the same
reasoning, a Pco 2 of 65 (resultant pH = 7.43), which is known to be beyond
the usual limit for compensation of a metabolic alkalosis, would suggest a
concomitant respiratory acidosis. Therefore, in evaluating the adequacy of
MIXED ACID BASE DISTURBANCES 351
respiratory compensation for a given metabolic alkalosis, it should be appre-
ciated that some degree of CO 2 retention is expected but that as the Pco 2 rises
above 55 mm Hg, a coexistent respiratory acidosis becomes progressively
more likely.
Respiratory disturbances are classified as acute or chronic depending on
the presence of renal compensation. The initial increase in serum HC0 3- in
respiratory acidosis or decrease in serum HC0 3- during respiratory alkalosis
occurs entirely through the action of body buffers which take up or release
hydrogen ions respectively in response to a change in carbonic acid (Le.,
Pco 2) excretion by the lungs. While this method of compensation by body
buffering is effective and rapid, the magnitude of the resultant change in
serum HC0 3- is limited to 3 to 4 mEq/L.2,6 Thus, in an acute respiratory
acidosis (elevated Pco 2), the serum HC0 3 should not increase above 28--30
mEq/L.6.9 Conversely, in a simple acute respiratory alkalosis (loweredPco 2 ),
the serum HCOi is usually not lower than 18 to 20 mEq/L.2 Respiratory
acidosis and alkalosis also elicit renal mechanisms of compensation but these
do not reach a maximum for 2 to 4 days, after which the disturbance can be
classified as "chronic." The kidney responds to a respiratory acidosis by in-
creasing net acid excretion thereby regenerating new bicarbonate which raises
the serum HC03 - level above that which occurs by body buffering alone. 3B
Conversely, during respiratory alkalosis, decreased renal acid excretion leads
to decreased bicarbonate regeneration, resulting in a decrease in serum
HC0 3- greater than that which occurs during acute respiratory alkalosis. 7, IS
As a clinical guide, it is useful to remember that the serum HC0 3- rises by
about 0.4 mEq/L per 1 mm Hg rise in Pco 2 in chronic respiratory acidosis 21 , 31
and is depressed by approximately 0.5 mEq/L per 1 mm Hg decrease in Pco 2
during chronic respiratory alkalosis. IS, 31 Again, there are limits to these
chronic changes. The serum HC0 3 is usually not increased beyond 45 mEq/L
in a compensated respiratory acidosis 37 , 3S or depressed below 15 mEq/L in a
compensated respiratory alkalosis. IS, 31 Knowledge of both the time course and
the magnitude of these compensatory changes for respiratory disturbances
can be used to identify a mixed disturbance. For example, in a patient with
known chronic respiratory alkalosis, a serum HC0 3- below 12 mEq/L would
indicate a superimposed metabolic acidosis since the HC0 3 level is too low to
be explained entirely by metabolic compensation for a respiratory alkalosis.
Similarly in a patient with long-standing CO 2 retention, a serum bicarbonate
of 50 mEq/L would suggest a concomitant metabolic alkalosis caused,
perhaps, to the excessive use of diuretics. To summarize, in assessing the
adequacy of compensation for a respiratory disturbance attention to the
historical facts to differentiate an acute from a chronic disturbance and
knowledge of the limits of compensation for each disorder to assess the blood
gas values frequently is all that is needed.

CLINICAL APPROACH TO MIXED DISTURBANCES

Dr. McCurdy, to whom this issue is dedicated, was noted for her teaching
of acid-base physiology. One of her greatest contributions to the subject was
the development of a systematic method for analyzing acid-base disorders
352 MARGARET BIA AND SAMUEL O. THIER

using a clinical approach. 28 The essential steps involved in this method are
outlined in Table 2. Basically, information from the clinical history, the
physical examination and the laboratory data are all utilized to identify an
acid-base disorder{s).
The first step in this approach is a careful history searching for clues that
lead the clinician to suspect the presence of certain acid-base disturbances.
Major causes for the simple acid-base disorders should be kept in mind so that
a disorder can be predicted by the presence of certain disease states. Thus, a
patient with chronic renal failure would be expected to have a metabolic
acidosis, while in a patient with a history of vomiting, a metabolic alkalosis
would be the predicted disorder. Acid-base disorders that are common in
certain clinical settings should also be remembered, such as respiratory
alkalosis in patients with pneumonia, sepsis or congestive heart failure, and
lactic acidosis in patients with septic shock. The ingestion of certain drugs
should alert the clinician to potential disturbances such as metabolic alkalosis
in a patient taking a loop-acting diuretic or metabolic acidosis in a subject
taking acetazolamide.
Additional clues to a possible acid-base disorder{s) can be obtained
during physical examination. For example, in a patient with clinical signs of
extracellular fluid volume contraction, a metabolic alkalosis should be sus-
pected. Similarly, tetany may be a manifestation of alkalemia; cyanosis may
reflect respiratory failure and therefore respiratory acidosis; fever is often
associated with respiratory alkalosis, etc. Thus, there are many signs on
physical examination which can lead the clinician to expect an acid-base
disturbance. This possibility is then either supported or excluded by an
evaluation of the laboratory data. Routine non-electrolyte chemistries are
useful in predicting disturbances. An elevated serum creatinine concentration
indicating renal failure makes a metabolic acidosis . likely. An increase in
serum glucose concentration with positive serum ketones suggests diabetic
ketoacidosis. Serum electrolytes provide further evidence for a possible
acid-base disorder.
When examining the serum electrolytes for signs of an acid-base disturb-
ance, four steps should be taken. First, the serum HC0 3 concentration should
be noted. Any deviation in serum HC0 3 from normal has one of two

Table 2. Clinical Approach to the Diagnosis


of Mixed Acid-Base Disturbances
I. Suspect the disturbances from the history
Il. Suspect the disturbauces from the physical examination
Ill. A. Evaluate routine laboratory data
B. Evaluate the electrolytes for
1. HCO, - if t , think of metabolic alkalosis or compensated respiratory acidosis
if ~ , think of metabolic acidosis or compensated respiratory alkalosis
2. K - if t , think of acidemia
if ~ , think of alkalemia
3. Cl- - if l' , think of hyperchloremic metabolic acidosis
if ~ , think of metabolic alkalosis
4. Anion gap
IV. Evaluate the blood gas values - check to see if compensation is appropriate.
Remember the limits of compensation.
MIXED ACID BASE DISTURBANCES 353
explanations. An elevation represents either a metabolic alkalosis or metabol-
ic compensation for a respiratory acidosis and a depression signifies either a
metabolic acidosis or metabolic compensation for a respiratory alkalosis.
Thus, unless the serum HC0 3 is abnormal beyond the limits of compensation
(Le., below 12 to 15 mEq/L for metabolic compensation of a respiratory
alkalosis or above 45 for metabolic compensation of a respiratory acidosis),
one cannot be sure of the correct disturbance from the serum bicarbonate
alone. However, a prediction can usually be made which is then confirmed by
measurement of blood pH.
Second, the serum potassium concentration should be examined. It is
well-known that the serum potassium concentration can be altered by
changes in acid-base balance such that acidemia can lead to hyperkalemia and
alkalemia may result in hypokalemia. 8,39 This relationship has been defined
by empirical observations indicating that a change in blood pH of 0.10 units
results in a change in the plasma potassium concentration of 0.6 mEq/L. In an
individual patient, however, there is considerable variation in this relation-
ship, with pH changes of 0.10 units being associated with changes in serum
potassium of 0.3 to 1.3 mEq/L.39 In addition, several factors can modify this
response. For instance, alterations in serum potassium are much more
pronounced with metabolic than with respiratory disturbances. 11. 18. 26. 38 Fur-
thermore, the fall in serum potassium concentration during metabolic alkalo-
sis is less than the rise that occurs with metabolic acidosis. 23 ,35 During
metabolic acidosis, the nature of the anion accompanying the increase in
hydrogen ion concentration has also been shown to influence the change in
serum potassium. Thus, in dogs, experimentally induced hyperchloremic
acidosis results in a significant elevation in serum potassium concentration
whereas infusion oflactic acid causes no change. 33 Whether the nature of the
anion is as critical in endogenous acidosis in human subjects is unknown.
Evidence also indicates that changes in serum HC0 3 alone, independent of
changes in blood pH, can influence the serum potassium levet 13 , 14,25 Lastly,
it has recently been reported in a series of patients with metabolic acidosis,
that the presence of hyperkalemia did not correlate with the degree of
acidemia, but rather was related to the presence of impaired renal function
and catabolism. 16 In summary, it is important to realize that the influence of
blood pH on changes in serum potassium concentration is complex and
cannot be defined by one equation. However, in examining a given set of
electrolytes, it is still useful to utilize the serum potassium level as an
indicator of a possible disturbance. Therefore, a low serum HC0 3 concentra-
tion associated with hyperkalemia can be used as evidence for a probable
metabolic acidosis while a high serum HC0 3 concentration in association
with hypokalemia most likely indicates a metabolic alkalosis.
Third, the serum chloride (Cl"':) should be noted since it is frequently
decreased in metabolic alkalosis whereas a high serum chloride can be
indicative of a hyperchloremic metabolic acidosis. However, during metabol-
ic compensation for respiratory disturbances, the change in renal hydrogen
ion excretion that occurs is associated with a change in chloride excretion 18 . 38
that can then result in an alteration in serum chloride level. For example, a
low serum HC0 3 in association with a high serum CI- may indicate either a
hyperchloremic metabolic acidosis or compensation for a respiratory alkalo-
sis.
354 MARGARET BIA AND SAMUEL O. TRIER

Fourth, a thorough examination of the electrolytes is not complete until


the anion gap is calculated. In normal subjects, the anion gap, which is usually
calculated as Na+ (CI- + HC03"), ranges between 8 and 16 mEq/L and is
composed mainly of unmeasured negatively charged proteins, phosphate,
sulfate, and organic ions. 12 , 32 The finding of an elevated anion gap can provide
important evidence for the presence of a metabolic acidosis even when there
is no decrease in serum HC0 3. The reader is referred to the article by Gardner
and Narins for a more detailed discussion of metabolic acidosis associated
with an increased anion gap. When analyzing the anion gap, other less
common causes for an increase in the amount of unmeasured anions should be
kept in mind. Infusion with a sodium salt of an unmeasured anion (e.g.,
sodium sulfate or sodium carbenicillin) can increase the gap.12 Severe
dehydration and metabolic alkalosis may also contribute to a widening of the
gap by increasing both the concentration and the negative charge of the
anionic plasma proteins that constitute the bulk of the normal anion gap;27,41
these processes can increase the anion gap by 5 to 9 mEq/L.17, 27 An anion gap
greater than 30 mEq/L virtually always indicates the presence of metabolic
acidosis, usually lactic or ketoacidosis. 17 In patients with an anion gap
between 20 and 30 mEq/L, an organic anion acidosis can frequently but not
always be identified. 17 With these limitations in mind, the anion gap remains a
valuable clinical tool in analyzing acid-base disturbances.
Frequently an adequate assessment of a patient's acid-base status can be
made from history, physical examination and the analysis of the electrolytes
described above. The decision to determine blood pH and Peo 2 depends on
the need to confirm the initial clinical impression; in very sick patients the
determination is usually essential. Measurement of pH and Peo 2 are also
necessary to document a mixed disturbance and to determine the extent of
acidemia or alkalemia upon which the need for treatment should be based. It
is important to realize that venous blood gases are often adequate for
determination of acid-base status and may, on occasion, be preferable (e.g. in
patients with a bleeding diathesis). In venous blood, Peo 2 is 5 to 7 mm
higher, HC0 3- 1 to 3 mEq/L lower, and pH 0.03 to 0.05 units lower than in an
arterial sample. 9 Since stasis will produce a local lactic acidosis, this measure-
ment requires the collection of free flowing venous blood obtained after the
tourniquet has been removed. However, in seriously ill patients and in other
clinical situations where it is critical to know the P0 2 of the patient, arterial
blood gases are required.
A careful analysis of the blood gas indices should begin with a check to
see if the values are consistent with the Henderson-Hasselbalch equation. If
not, one or more of the measurements must be inaccurate. Reference to a
blood gas map19 where the relationship between pH, Peo and HC03 as
defined by the Henderson-Hasselbalch equation is plotted, is one way to
check the validity of the numbers. Alternatively, the internal consistency of
the values can be verified by utilizing the reported Pco 2 and HC0 3- to
calculate the expected hydrogen ion concentration according to Equation 2 as
described above. The next step is to examine the pH, Pco 2 and HC0 3- to
determine if the values are indicative of a simple or a mixed disturbance.
Frequently, a mixed acid-base disturbance is already suspected from an
analysis of the clinical data and is confirmed when inappropriate compensa-
MIXED ACID BASE DISTURBANCES 355
tion for the primary or most highly suspected disorder is demonstrated. Thus,
when the PC0 2 is above or below the predicted value for compensation of a
metabolic disturbance or when the HC0 3 - level is above or below the level
expected for compensation of a given respiratory disorder, a mixed disturb-
ance is present. For this purpose, the rules of compensation can be remem-
bered or can be referred to from an easily retrievable source. The formula
described above in Equation 3 should be used when evaluating the adequacy
of ventilation for a metabolic acidosis. 1 For other disturbances, however, it is
often more practical to recall the limits of compensation, as)isted in Table 1.
These limits can then be applied to determine the appropriateness of
compensation. The time course of the compensatory changes must also be
remembered so that mixed disturbances are not over diagnosed in situations
where compensatory processes have not had enough time to reach a maximum
or, conversely, to resolve once the primary disturbance is corrected.
A few final points are useful to remember when analyzing the pH, Pco 2
and HC0 3 - determinations. In the setting of a known primary disorder, the
presence of a normal pH usually implies a mixed disturbance since compen-
sation rarely corrects pH back to normal (except in some cases of chronic
respiratory alkalosis). Generally, the more marked the primary disturbance,
the less likely the pH will be normal unless there is a mixed disorder.
Obviously, the finding of a pH change in a direction opposite to that predicted
for a known primary disorder demands the diagnosis of a mixed disturbance.
It should also be remembered that in a simple disturbance, Pco 2 and HC0 3 -
are always deviated in the same direction. Therefore, if the Pco 2 and HC0 3 -
are altered in opposite directions, a mixed disturbance must be present. In
summary, then, the pH, Pco 2 , and HC0 3 - should always be analyzed to
determine the adequacy of compensation for the primary disturbance. There
are then several clues which, when combined with the clinical data, can
provide the evidence for a mixed acid-base disturbance. Finally, reference to
an acid-base nomogram may be helpful in confirming the presence of a simple
disorder if the values fall within the confidence band of a single disturbance.
However, the fact that the values fall within a single band does not alone
prove that there is only one disorder but merely provides support for this
possibility which can then be considered verified if consistent with the rest of
the clinical data.

TREATMENT

The common mixed acid-base disturbances and examples of some clinical


settings in which these combinations occur are listed in Table 3. Simultane-
ously occurring respiratory plus metabolic acidosis or respiratory plus meta-
bolic alkalosis produce the most severe change in blood pH since the
respiratory disturbance adds to rather than compensates for the metabolic
disorder and vice versa. When making therapeutic decisions about mixed
acid-base disturbances, there are two important principles to recall. First, the
aim of treatment is to return the blood pH toward normal. Therefore, it is the
pH which must be followed to determine when to begin or to discontinue
therapy. For example, bicarbonate treatment might be administered to a
356 MARGARET BIA AND SAMUEL O. THIER

Table 3. Important Causes of Mixed Acid-Base Disturbances


1. Respiratory acidosis with metabolic acidosis
Example: Cardiopulmonary arrest
Severe pulmonary edema
Drug ingestion with central nervous system depression

2. Respiratory alkalosis and metabolic alkalosis


Example: Hepatic failure and diuretics
Patients on ventilation given nasogastric suction

3. Respiratory alkalosis and metabolic acidosis


Example: Septic shock
Renal failure with sepsis
Salicylate overdose

4. Respiratory acidosis and metabolic alkalosis


Example: Chronic lung disease and diuretic use

5. Mixed acute and chronic respiratory acidosis


Example: Chronic lung disease and superimposed infection

6. Metabolic acidosis and metabolic alkalosis


Example: Renal failure and vomiting
Vomiting and hypotension (lactic acidosis)

patient with a serum HCO a- of 11 mEq/L and a pH of 7.20 but should not be
given to a patient with a serum HCO a- of 11 mEq/L but with a normal pH of
7.40 because of a concomitant primary respiratory alkalosis. This is not to
imply that the acid-base disturbances in the second patient is less severe than
in the first patient but only to emphasize that the need for specific therapy
depends on the blood pH. Second, in treating combined disorders, the effect
of treatment of one disturbance on the manifestation of the second disorder
should be anticipated. For example, in a patient with a chronic respiratory
acidosis plus a metabolic alkalosis, improving ventilation to lower Pco 2
without simultaneously treating the alkalosis to lower HCO a- can'leave the
patient with a significant alkalemia. Thus the goal of therapy in mixed
disorders is to treat the blood pH, always being aware of the effect of such
therapy on all of the disturbances present.
While there are a large number of possible combinations, a few mixed
disturbances deserve special mention because of their clinical importance;
these are listed in Table 3. No attempt has been made to review all mixed
disturbances but rather to describe the ones that are most likely to be
encountered clinically and are important because of their diagnostic and
therapeutic implications.
Respiratory Acidosis Plus Metabolic Acidosis
This combination occurs in a variety of clinical situations including
cardiopulmonary arrest, severe pulmonary edema, drug ingestion with severe
central nervous system depression and hypoventilation, and metabolic acido-
sis with potassium depletion producing paralysis of the respiratory muscles.
In these cases, the serum bicarbonate is usually low, the Pco 2 normal or
MIXED ACID BASE DISTURBANCES 357
elevated, and the resultant pH may be dangerously low. The CO 2 retention
prevents respiratory compensation for the metabolic acidosis and the meta-
bolic process prevents compensation for the respiratory acidosis. Significant
acidemia can seriously impair cardiac function and lead to cardiovascular
collapse. 3o Specific therapy must be initiated aggressively to correct the
acidemia by simultaneously treating the metabolic acidosis with bicarbonate
and the respiratory acidosis with measures to improve ventilation. Frequently
intubation is required in these situations, a fact which may be obvious when
the Pco 2 is 60 mm Hg but may not be so easy to appreciate if the Pco 2 is not
elevated much above 40 mm Hg. However, as discussed earlier, a normal
Pco 2 in the setting of a severe metabolic acidosis may be as strong evidence of
respiratory acidosis as is a Pco 2 of 60 mm Hg in a patient without a metabolic
acidosis. The clinician should also be aware that the sodium bicarbonate
therapy for the metabolic acidosis may cause excessive volume expansion and
further compromise ventilation. Rules for calculating the amount ofbicarbon-
ate necessary are described by Drs. Gardner and Narins. A final point to
remember regarding treatment for a mixed respiratory and metabolic acidosis
is that hyperkalemia is frequently a serious problem in this setting requiring
aggressive therapy both to shift potassium intracellularly and to remove it
from the body.
Respiratory Alkalosis Plus Metabolic Alkalosis
This mixed disorder is commonly present in patients with hepatic failure
who are placed on diuretics or nasogastric suction. It is also common in
critically ill patients who require ventilator support and are given diuretics or
nasogastric suction. 43 In these cases, the serum HC03 - is usually elevated, the
Pco 2 is normal or low and the blood pH is, of course, alkaline. This disorder
deserves special mention because the profound alkalemia which may result
demands immediate specific therapy. Severe alkalemia can adversely affect
both cerebral and peripheral hemodynamics. 4 , 24, 30 and is associated with a
poor prognosis in critically ill patients. 43 In this mixed disturbance, the
metabolic process prevents compensation for the respiratory alkalosis and the
hyperventilation prevents compensation for the metabolic alkalosis. In order
to return the pH toward normal, therapy theoretically should again be
directed at alleviating both disorders simultaneously. Treatment of the
metabolic alkalosis with volume, chloride and potassium replacement should
be initiated. If the patient is on a ventilator, the settings should be adjusted to
increase the Pco 2 • However, raising the Pco 2 is more difficult in a patient
with spontaneous hyperventilation unless the underlying process leading to
the increased respiratory drive can be corrected. In these cases, the clinician
should accept the presence of the respiratory alkalosis and concentrate efforts
on reversing the metabolic process. Potassium deficits in this setting may be
significant and replacement with large amounts of the cation should be an-
ticipated.
Respiratory Alkalosis Plus Metabolic Acidosis
Clinical settings in which this combination can be found include septic
shock, pulmonary embolism, renal failure with sepsis and salicylate ingestion.
The metabolic acidosis in these cases is frequently of the anion gap variety
358 MARGARET BIA AND SAMUEL O. THIER

(e.g., lactic acidosis, retention of organic anions in renal failure or salicylate


ingestion). The serum HC0 3 - is usually markedly decreased and the Pco 2
independently depressed beyond the limits predicted for respiratory compen-
sation of a metabolic acidosis. The resultant pH may be normal or just mildly
deviated from normal depending on the severity of the individual disturb-
ances. Examining the respiratory compensation for the metabolic acidosis is
critical in this situation since, frequently, the respiratory alkalosis will be
missed until it is realized that ventilation is greater than that predicted by
compensatory responses alone. Specific treatment aimed at correcting the pH
is often not necessary. In fact, bicarbonate therapy may be contraindicated
since, in the presence of a severe respiratory alkalosis, it could produce or
exacerbate alkalemia. Therefore, the main value in recognizing this disturb-
ance is more for its diagnostic potential than for the purpose of instituting
specific therapy. Treatment should be directed at correcting the underlying
processes.
Respiratory Acidosis Plus Metabolic Alkalosis
Although this combination can occur in several situations, it is present
most commonly in patients with chronic lung disease and CO 2 retention. 3 , 29, 37
In these patients, a metabolic alkalosis frequently arises because of vomiting
or treatment with diuretics and a low salt diet. 3 ,37 The Pco 2 is elevated, the
serum HC0 3 - is elevated even more than is expected for metabolic compen-
sation for a respiratory acidosis and the resultant pH may be normal or mildly
depressed. Therefore, it is often not necessary to institute specific measures to
correct the blood pH. It is important, however, to recognize and treat the
primary metabolic alkalosis with volume, chloride and potassium replace-
ment since the elevated bicarbonate may itself depress respiration even
further.29 As mentioned earlier, this fact is especially important when attempt-
ing to wean a patient with chronic CO 2 retention from the respirator.
Mixed Acute and Chronic Respiratory Acidosis
This combined disturbance typically occurs in a patient with chronic
respiratory insufficiency and CO 2 retention. The retained CO 2 pr~duces a
renal response that results in an appropriate compensatory elevation of serum
HC0 3 • The patient then develops a superimposed impairment in respiratory
function resulting, for instance, from pulmonary infection which leads to
further CO 2 retention. Thus, a patient with a baseline Pco 2 of 55 and HC0 3 -
of 30 (pH = 7.36) could then present with a Pco 2 of 75 and an HC0 3 - of 33
(pH = 7.27). That an acute respiratory acidosis is superimposed on a chronic
respiratory acidosis is suggested from the clinical history. This impression is
then confirmed by the blood gas parameters showing a Pco 2 of 75, which is
higher than would be tolerable on a chronic basis and therefore implies an
acute change, with a HCO a of 33 which represents evidence of chronic
exposure to an elevated Pco 2. Classification of this clinical disorder as amixed
disturbance is appropriate since acute and chronic respiratory acidosis are
considered as separate disturbances because of the time needed to convert an
acute to a chronic respiratory acidosis via renal compensation. Obviously the
treatment for this combined respiratory acidosis should be directed at improv-
ing ventilatory function to lower the Pco 2. Ventilatory support may be
necessary until infection or other precipitating event can be reversed.
MIXED ACID BASE DISTURBANCES 359
Metabolic Alkalosis and Metabolic Acidosis
This mixed disorder is seen frequently when patients with long-standing
metabolic acidosis from chronic renal failure or diarrhea then develop a
metabolic alkalosis from vomiting. Alternatively, the combination can occur
in patients with an initial metabolic alkalosis from severe vomiting who then
develop sufficient volume depletion to result in tissue hypoperfusion and a
superimposed lactic acidosis. The HC0 3 and Pco 2 may be high, low, or
normal depending on the intensity of each disturbance and the resultant pH is
often close to normal. An increase in the anion gap may be the clue to the
presence of the metabolic acidosis in these situations. Since pH may be little
altered from normal, treatment should be directed at resolving the underlying
processes, always remembering that the resolution of one disturbance will
permit the second abnormality to emerge unopposed.

SUMMARY

The analysis of a mixed acid-base disturbance begins with the history and
physical examination from which data can be derived that make the clinician
suspect a specific disturbance(s). The electrolytes are then evaluated with
emphasis on the meaning of the values for serum bicarbonate, potassium and
chloride concentration and on the level of the anion gap. Other laboratory data
such as serum creatinine or glucose concentrations, blood cultures, and so
forth, should also be reviewed for further clues to a possible disturbance(s).
When it is clinically indicated, values for pH and Pco 2 are obtained by
blood gas determination. If the evidence up to this point indicates the
presence of at least one disturbance, the data are examined to see if
compensation for this disturbance is appropriate. If not, a mixed disturbance
must be present. A normal pH in the setting of an abnormal serum HC0 3 -
concentration or Pco 2 also suggests a mixed disturbance since compensation
rarely corrects the pH back to normal. Of course, a pH deviated in the
opposite direction than that expected' for a known primary disturbance makes
the diagnosis of a mixed disturbance certain.
The diagnosis of a mixed acid-base disturbance is therefore based on an
analysis of all the clinical data and not just the blood gas measurements.
Treatment of the disorders should be directed at maintaining a normal or near
normal pH. Some combined acid-base disorders are important to recognize
because they can result in a severe deviation in blood pH that demands
immediate, specific therapy. Other mixed disturbances result in a pH which is
near normal but are important to recognize since they can alert the clinician to
the possibility of certain clinical derangements such as septic shock or drug
ingestion. Careful analysis of mixed acid-base disturbances in this way is not
difficult and can provide the clinician with important diagnostic and thera-
peutic information to be used in caring for his (her) patients.

ACKNOWLEDGMENT

We are indebted to Dr. John Forrest for his helpful comments and sug-
gestions.
360 MARGARET BIA AND SAMUEL O. THIER

REFERENCES
1. Albert, M.D., Dell, R B., and Winters, R W.: Quantitative displacement of acid base
equilibrium in metabolic acidosis. Ann. Intern. Med., 66:312, 1964.
2. Arbus, G. S., Herbert, L. A., Levesque, P. R., et al.: Characterization and clinical application of
the "significance band" for acute respiratory alkalosis. New Engl. J. Med., 280: 117, 1969.
3. Bear, R, Goldstein, M., Phillipson, E., et al.: Effect of metabolic alkalosis on respiratory
function in patients with chronic obstructive lung disease. Can. Med. Assoc. J., 117:900,
1977.
4. Betz, E., and Heuser, D.: Cerebral cortical blood flow during changes of acid-base equilibrium
of the brain. J. Appl. Physiol., 23:726, 1976.
5. Bone, J. M., Cowie, J., Lambie, A., et al.: The relationship between arterial Peo, and hydrogen
ion concentration in chronic metabolic acidosis and alkalosis. Clin. Sci. Mol. Med., 46:113,
1974.
6. Brackett, N. C., Cohen, J. J., and Schwartz, W. B.: Carbon dioxide titration curve of normal man:
Effect of increasing degrees of acute hypercapnia on acid-base equilibrium. New Engl. J.
Med., 272:6, 1965.
7. Brown, E. B.: Electrolyte changes with chronic passive hyperventilation in man. J. App.
Physiol., 1 :848, 1949.
8. Burnell, J. M., Villamel, M. F., Uyeno, B. T., et al.: Effect in humans of extracellular pH changes
in relationship between serum potassium concentration and intracellular potassium. J. Clin.
Invest., 35:935, 1956.
9. Cohen, J. J., and Kassirer, J. P.: Acid base metabolism. In Maxwell, M. H., and Kleeman, C. R.,
eds.: Clinical Disorders of Fluid and Electrolyte Metabolism. New York, McGraw-Hill, 1980,
pp. 181-232.
10. Elkington, J. R: Clinical disorders of acid-base regulation. A survey of seventeen years'
diagnostic experience. Med. Clin. N. Amer., 50:1325, 1966.
11. Elkington, J. R, Singer, R B., Barker, E. S., et al.: Effects in man of acute respiratory alkalosis
and acidosis on ion transfer in the total body fluids. J. Clin. Invest., 34:1671, 1955.
12. Emmett, M., and Narins, R G.: Clinical use of the anion gap. Medicine, 56:38,1977.
13. Fraley, D. S., and Adler, S.: Isohydric regulation of plasma potassium by bicarbonate in the rat.
Kidney Internat., 9:333, 1976.
14. Fraley, D. S., and Adler, S.: Correction ofhyperkalemia by bicarbonate despite constant blood
pH. Kidney Internat., 12:354, 1977.
15. Fulop, M.: Hypercapnia in metabolic alkalosis. N.Y. State J. Med., 76:19, 1976.
16. Fulop, M.: Serum potassium in lactic acidosis and ketoacidosis. New Engl. J. Med., 300:1087,
1979.
17. Gabow, P. A., Kaehny, W., Fennessey, P. V., et al.: Diagnostic importance of an increased serum
anion gap. New Engl. J. Med., 303:854, 1980.
18. Gennari, F. J., Goldstein, M. B., and Schwartz, W. B.: The nature of the renal adaptation to
chronic hypocapnia. J. Clin. Invest., 51: 1722, 1972.
19. Goldberg, M., Green, S. B., Moss, M. L., et al.: Computerbased instruction and diagnosis of
acid-base disorders. J.A.M.A., 223:269, 1973.
20. Goldring, R M., Cannon, P. J., Heinemann, H. 0., et al.: Respiratory adjustment"to chronic
metabolic alkalosis in man. J. Clin. Invest., 47:188, 1968.
21. Kaehny, W. D.: Pathogenesis and management of respiratory and mixed acid-base disorders. In
Schrier, R W., ed.: Renal and Electrolyte Disorders. Boston, Little, Brown and Co., 1976, pp.
121-142.
22. Kassirer, J. P., and Bleich, H. L.: Rapid estimation of plasma carbon dioxide from pH and total
carbon dioxide content. New Engl. J. Med., 272:1067, 1965.
23. Keating, R. E., Weichselbaum, T. E., Alanis, M., et al.: The movement of potassium during
experimental acidosis and alkalosis in nephrectomized dog. Surg. Gynec. Obstet., 96:323,
1953.
24. Kilburn, K. H.: Shock, seizures and coma with alkalosis during mechanical ventilation. Ann.
Intern. Med., 65:979, 1966.
25. Kim, W. G., and Brown, E. B.: Potassium transfer with constant extracellular pH. J. Lab. Clin.
Med., 71 :678, 1968.
26. Liebman, J., and Edelman, I. S.: Interrelationship of plasma potassium concentration, plasma
sodium concentration, arterial pH and total exchangeable potassium. J. Clin. Invest., 38:2176,
1959.
27. Madias, N. E., Ayus, J. C., and Androgue, H. J.: Increased anion gap in metabolic alkalosis: The
role of plasma protein equivalency. New Engl. J. Med., 300:1421, 1979.
28. McCurdy, D. K.: Mixed metabolic and respiratory acid-base disturbances: Diagnosis and
treatment. Chest, 62:35S, 1972.
29. Miller, P. D., and Berns, A. S.: Acute metabolic alkalosis perpetuating hypercarbia. A role for
acetazolamide in chronic obstructive pulmonary disease. J.A.M.A., 238:2400, 1977.
30. Mitchell, J. H., Wildenthal, K., and Johnson, R L.: The effects of acid-base disturbances on
cardiovascular and pulmonary function. Kidney Internat., 1 :375, 1972.
MIXED ACID BASE DISTURBANCES 361
31. Narins, R G., and Emmett, M.: Simple and mixed acid-base disorders: a practical approach.
Medicine, 59:161,1980.
32. Oh, M. S., and Carroll,H. J.: The anion gap. New Engi. J. Med., 297:814,1977.
33. Oster, J. R, Perez, G. 0., and Vaamonde, C. A.: Relationship between blood pH and potassium
and phosphorus during acute metabolic acidosis. Amer. J. Physioi., 235:F345, 1978.
34. Pierce, N. F.: The ventilatory response to acute base deficit in humans. Ann. Intern. Med.,
72:633, 1970.
35. Poole, P. A., and Cameron, 1. R: Intracellular pH and K+ of cardiac and skeletal muscle in
acidosis and alkalosis. Amer. J. Physioi., 229:1305,1975.
36. Relman, A. S.: Metabolic acidosis. Med. Times, 96:1094, 1968.
37. Robin, E. D.: Abnormalities of acid base regulation in chronic pulmonary disease with special
reference to hypercapnia and extracellular alkalosis. New Engi. J. Med., 268:917,1963.
38. Schwartz, W. B., Brackett, N. C., and Cohen, J. J.: The response of extracellular hydrogen ion
concentration to graded degrees of chronic hypercapnia: The physiologic limits of the defense
of pH. J. Clin. Invest., 44:291, 1965.
39. Simmons, D. H., and Avedon, M.: Acid-base alterations and plasma potassium concentration.
Amer. J. Physioi., 197:319, 1959.
40. Statement on acid-base terminology. Ann. Intern. Med., 63:885, 1965.
41, Van Slyke, D. D., Hastings, A. B., HilIer, A., et ai.: Studies of gas and electrolyte equilibria in
blood. XIV. The amounts of alkali bound by serum albumin and globulin. J. Bioi. Chem.,
79:769, 1928.
42. Van Ypersele de Strihou, C., Brasseur, L., and DeConinck, J.: The "carbon dioxide curve" for
chronic hypercapnia in man. New Engi. J. Med., 275:117,1966.
43. Wilson, R F., Gibson, D., Percinel, A. K., et al.: Severe alkalosis in critically ill surgical patients.
Arch. Surg., 105:197, 1972.

Yale-New Haven Hospital


333 Cedar Street
New Haven, Connecticut 06510

You might also like